CN103886622B - The implementation method of automated graphics region division and realize device - Google Patents
The implementation method of automated graphics region division and realize device Download PDFInfo
- Publication number
- CN103886622B CN103886622B CN201210560218.3A CN201210560218A CN103886622B CN 103886622 B CN103886622 B CN 103886622B CN 201210560218 A CN201210560218 A CN 201210560218A CN 103886622 B CN103886622 B CN 103886622B
- Authority
- CN
- China
- Prior art keywords
- pixel
- main characteristic
- characteristic vector
- vector
- accumulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of implementation method of automated graphics region division, it includes:Obtain the smooth structure tensor field corresponding to each pixel in image;The degrees of detail of described image is obtained according to the accumulated change degree of the smooth structure tensor field;Region division is carried out to described image according to the degrees of detail of described image.What the present invention also provided a kind of automated graphics region division that can realize the above method realizes device.The implementation method and device for the automated graphics region division that the present invention is provided do not need any user mutual to complete the region division of image, calculate relatively easy, the region division of image can also be realized on the platform of poor-performing.
Description
Technical field
The present invention relates to digital image processing techniques, more particularly to a kind of automated graphics region based on image detail degree
The implementation method of division and realize device.
Background technology
In the research and application of image, people are often only interested in some of piece image part, and these senses are emerging
The part of interest, which generally corresponds to region specific, with special nature in image, (can correspond to single region, can also correspond to many
Individual region), referred to as target or prospect;And other parts are referred to as the background of image.In order to recognize and analyze target, it is necessary to mesh
Mark isolates out from piece image, here it is the problem of image segmentation will be studied.So-called image segmentation, in a larger sense,
It is that image pixel is entered according to the similarity criterion of some features or characteristic set of image (including gray scale, color, texture etc.)
Row grouping and clustering, the plane of delineation, which is divided into several, has the not overlapping region of some uniformity.This causes in the same area
Pixel characteristic be similar, i.e., with uniformity;And the feature of pixel has mutation between different zones, i.e., with non-uniform
Property.
Existing technical scheme is general by the way of demarcation by hand when distinguishing different zones to image, automatically side
Method is usually classical Graph Cut scheduling algorithms, by with user mutual, to the statistical nature of the image in user's selection area
Analyzed, then the front and rear scape of image is made a distinction automatically.
Existing technology, such as using the method in demarcation region by hand, it is clear that again inaccurate during operating cost, and Graph cut
User's selected object and background are required for when handling each width picture etc. classic algorithm, it is time-consuming longer, calculate also complex,
It is then to need to expend more times on the platform that mobile phone etc. calculates poor-performing.
The content of the invention
It is an object of the present invention to overcome the defect present in the technology of existing image-region division, and provide a kind of
The implementation method of new automated graphics region division, it can quickly be calculated in the case of completely without user mutual
Quick image-region division is carried out to the degrees of detail index of each pixel of image, and according to this index.
The object of the invention to solve the technical problems is realized using following technical scheme.
The present invention provides a kind of implementation method of automated graphics region division, and it includes:Obtain each pixel in image
Corresponding smooth structure tensor field;The details of described image is obtained according to the accumulated change degree of the smooth structure tensor field
Degree;Region division is carried out to described image according to the degrees of detail of described image.
What the present invention provided a kind of automated graphics region division realizes device, and it includes:Smooth structure tensor field obtains mould
Block, for obtaining the smooth structure tensor field in image corresponding to each pixel;Image detail degree acquisition module, according to described
The accumulated change degree of smooth structure tensor field obtains the degrees of detail of described image;Image-region division module, according to the figure
The degrees of detail of picture carries out region division to described image.
The automated graphics region division that the present invention is provided realizes that device does not need any user mutual to complete image
Region division, calculate it is relatively easy, the region division of image can also be realized on the platform of poor-performing.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, and in order to allow the above and other objects, features and advantages of the present invention can
Become apparent, below especially exemplified by preferred embodiment, and coordinate accompanying drawing, describe in detail as follows.
Brief description of the drawings
Fig. 1 be first embodiment of the invention in automated graphics region division implementation method schematic flow sheet.
Fig. 2 is a sub-picture of pending region division.
Fig. 3 is the specific schematic flow sheet of the step S11 shown in Fig. 1.
Fig. 4 is the idiographic flow schematic diagram of the step S12 shown in Fig. 1.
Fig. 5 be first embodiment of the invention in calculate accumulation direction value of each pixel along its main characteristic vector positive direction
Idiographic flow schematic diagram.
Fig. 6 be first embodiment of the invention in calculate accumulation direction value of each pixel along its main characteristic vector opposite direction
Idiographic flow schematic diagram.
Fig. 7 is the degrees of detail figure that Fig. 2 is obtained after step S11 and S12 processing.
Fig. 8 is region division results of the Fig. 7 after step S13 processing.
Fig. 9 be second embodiment of the invention in automated graphics region division the structural representation for realizing device.
Figure 10 is the structural representation of the smooth structure tensor field acquisition module 21 shown in Fig. 9.
Figure 11 is the structural representation of the image detail degree acquisition module 22 shown in Fig. 9.
Figure 12 is positive direction accumulation direction value acquisition module 222a structural representation.
Figure 13 is opposite direction accumulation direction value acquisition module 222b structural representation.
Embodiment
Further to illustrate the present invention to reach the technological means and effect that predetermined goal of the invention is taken, below in conjunction with
Accompanying drawing and preferred embodiment, to according to the implementation method of automated graphics region division proposed by the present invention its embodiment,
Method, step, structure, feature and its effect, are described in detail as follows.
For the present invention foregoing and other technology contents, feature and effect, in the following preferable reality coordinated with reference to schema
Applying in the detailed description of example to clearly appear from.By the explanation of embodiment, when predetermined mesh can be reached to the present invention
The technological means taken and effect be able to more deeply and it is specific understand, but institute's accompanying drawings are only to provide with reference to saying
It is bright to be used, not for being any limitation as to the present invention.
First embodiment
Fig. 1 is the schematic flow sheet of the implementation method of disclosed automated graphics region division.As shown in Fig. 1,
The implementation method of the automated graphics region division of the present invention includes:
S11:Obtain the smooth structure tensor field corresponding to each pixel in image.
In this step image only refer to original image without any processing, refer to Fig. 2, be to treat shown in Fig. 2
Carry out a sub-picture of region division.
Fig. 3 is refer to, in step S11, the smooth structure tensor field corresponding to each pixel in image is obtained
Specific method may comprise steps of:
S111:Processing is filtered to described image, the gradient of each pixel is obtained.
Each gradient of the pixel on x, y directions in image can be for example calculated by sobel operators, specific formula is such as
Under:
Wherein, R, G, B represent it is each pixel corresponding red (R), green (G), the component of blue (B), f respectively
For x, the gradient vector on y directions.
Step S111 purpose is to calculate x respectively, the gradient on y directions, so any can calculate x, y side
The algorithm of upward gradient can be used, however it is not limited to calculate gradient by sobel operators.
S112:According to its corresponding tensor field of the gradient calculation of each pixel.
In step S112, according to the tensor field of each pixel of gradient calculation of each pixel on x, y direction, tool
Body formula is as follows:
S113:Tensor field to each pixel is smoothed the acquisition smooth structure tensor field.
In step S113, the tensor field to each pixel does smoothing processing, such as tensor field to each pixel
Do Gaussian Blur processing.Tensor field after Gaussian Blur can be usedRepresent.It is of course also possible to use other smooth
Processing mode, such as average be fuzzy etc., and processing mode does smoothing processing to the tensor field of each pixel, and the present invention is not with this
It is limited.
S12:The degrees of detail of described image is obtained according to the accumulated change degree of the smooth structure tensor field.
Fig. 4 is refer to, in step S12, further be may comprise steps of:
S121:According to the main characteristic vector of each pixel of smooth structure tensor field computation of each pixel.
According to the smooth structure tensor field computation characteristic vector of each pixel, specific formula is as follows:
Wherein, v1, v2 are the characteristic vectors of the pixel, and v1 is the main characteristic vector of the pixel.
S122:Accumulated change journey for characterizing the smooth structure tensor field is obtained according to the main characteristic vector of pixel
The accumulation direction value of the main characteristic vector of each pixel of degree.
In step S122, accumulation direction value of each pixel along its main characteristic vector positive direction is calculated respectively and every
Accumulation direction value of the individual pixel along its main characteristic vector opposite direction.
Fig. 5 is refer to, calculating the method for accumulation direction value of each pixel along its main characteristic vector positive direction can wrap
Include following steps:
Step a1:The pixel of accumulation direction value of main characteristic vector to be calculated is set as the first pixel P, chooses described
In image positioned at first pixel P main characteristic vector positive direction and be unit distance with first pixel P distance
Second pixel Q, obtains the primary vector A that the first pixel P and second pixel Q is constituted;
Step b1:Choose the main characteristic vector positive direction that is located at second pixel Q in described image and apart from for unit
3rd pixel Q1 of distance, obtains the secondary vector B that the second pixel Q and the 3rd pixel Q1 is constituted;
Step c1:The angle α i for calculating and storing primary vector A between secondary vector B, and record execution the step
Rapid cumulative frequency i;
Step d1:Judge whether the cumulative frequency is less than pre-determined number, if it is judged that being yes, then by second pixel
Point Q is set as the first pixel P (namely making P=Q), the 3rd pixel Q1 is set as to the second pixel Q (namely Q
=Q1), jump procedure b1 is otherwise, all stored angles of acquisition and square along its main characteristic vector as the pixel
To accumulation direction value.So that pre-determined number is 5 times as an example, then accumulation direction of the pixel along its main characteristic vector positive direction
Value=α 1+ α 2+ α 3+ α 4+ α 5.Unit distance can be set as a pixel or two pixels etc. according to being actually needed.
Fig. 6 is refer to, similarly, the method for calculating accumulation direction value of each pixel along its main characteristic vector opposite direction can
To comprise the following steps:
Step a2:The pixel of accumulation direction value of main characteristic vector to be calculated is set as the first pixel P, chooses described
In image positioned at first pixel P main characteristic vector opposite direction and be unit distance with first pixel P distance
Second pixel Q ', obtain the primary vector A ' that the first pixel P and the second pixel Q ' are constituted;
Step b2:Choose the main characteristic vector opposite direction that is located at the second pixel Q ' in described image and apart from for unit
3rd pixel Q1 ' of distance, obtain the secondary vector B ' that the second pixel Q ' and the 3rd pixel Q1 ' are constituted;
Step c2:The angle β i for calculating and storing primary vector A ' between secondary vector B ', and record execution should
The cumulative frequency i of step;
Step d2:Judge whether the cumulative frequency is less than pre-determined number, if it is judged that being yes, then by second pixel
Point is set as the first pixel, the 3rd pixel is set as to the second pixel, performs step b2, otherwise, obtains all quilts
Angle and as the pixel along its main characteristic vector opposite direction the accumulation direction value of storage.Calculating each pixel edge
The pre-determined number of step c1 and step c2 cumulative frequency is identical during the accumulation direction value of its main characteristic vector positive and negative direction
, the length of unit distance is also identical.For example calculating accumulation side of each pixel along its main characteristic vector positive direction
During to value, it is 5 times to set pre-determined number, then calculating accumulation direction value of each pixel along its main characteristic vector opposite direction
When, the pre-determined number of setting is also 5 times.Equally, so that pre-determined number is 5 times as an example, then the pixel is along its main characteristic vector
The accumulation direction value of opposite direction=β 1+ β 2+ β 3+ β 4+ β 5.
Cumulative (the α 1+ α 2+ α 3+ α 4+ α 5+ β of each accumulation direction value of the pixel along its main characteristic vector positive and negative direction
1+ β 2+ β 3+ β 4+ β 5) be exactly the corresponding main characteristic vector of the pixel accumulation direction value.
S123:The accumulation direction value of the main characteristic vector of all pixels point is normalized and reverse process obtains each
The degrees of detail of pixel.
The accumulation direction value of the main characteristic vector of all pixels point is normalized each picture of described image namely
The maximum of the accumulation direction value of the accumulation direction value of the main characteristic vector of vegetarian refreshments divided by the main characteristic vector of all pixels point.Will
Accumulation direction value after normalization reversely refers to subtracting the accumulation direction value after normalization, such as accumulation after normalizing with 1
Direction value a is the number in the range of 0~1, and the meaning reverse to a is exactly to make a=1-a.
The accumulation direction value of the main characteristic vector of all pixels point is normalized and reverse process obtains characterizing
The degrees of detail index of each pixel of image detail degree, i.e., the degrees of detail of each pixel.Fig. 7 is refer to, Fig. 7 is Fig. 2
The degrees of detail figure obtained after step S11 and S12 processing.
S13:Region division is carried out to described image according to the degrees of detail of described image.
After the degrees of detail for obtaining image, the mode that can be divided with classical threshold value carries out the division of degrees of detail to image,
Finally give the different degrees of detail regions of image.Fig. 8 is refer to, Fig. 8 is region division knots of the Fig. 7 after step S13 processing
Really, wherein, white portion be degrees of detail be less than threshold value low degrees of detail region.
The implementation method for the automated graphics region division that the present embodiment is provided does not need any user mutual to complete figure
The region division of picture, calculates relatively easy, the region division of image can also be realized on the platform of poor-performing.
Second embodiment
Fig. 9 is the structural representation for realizing device of disclosed automated graphics region division.Such as Fig. 9 institutes
Show, automated graphics region division of the invention realizes that device 20 includes:Smooth structure tensor field acquisition module 21, image detail
Spend acquisition module 22, image-region division module 23.
Smooth structure tensor field acquisition module 21 is used to obtain the smooth structure tensor in image corresponding to each pixel
;Image detail degree acquisition module 22 obtains the details of described image according to the accumulated change degree of the smooth structure tensor field
Degree;Image-region division module 23 carries out region division according to the degrees of detail of described image to described image.
Figure 10 is refer to, the smooth structure tensor field acquisition module 21 can further include:Gradient extraction module
211st, tensor field acquisition module 212, smoothing module 213.
Gradient extraction module 211 is used to be filtered described image processing, obtains the gradient of each pixel;Tensor field
Acquisition module 212 is used for according to its corresponding tensor field of the gradient calculation of each pixel;Smoothing module 213 is used for every
The tensor field of individual pixel is smoothed the acquisition smooth structure tensor field.
Figure 11 is refer to, described image degrees of detail acquisition module 22 can further include:Main characteristic vector computing module
221st, accumulation direction value acquisition module 222, degrees of detail acquisition module 223.
Main characteristic vector computing module 221 is used for each pixel of smooth structure tensor field computation according to each pixel
Main characteristic vector;Accumulation direction value acquisition module 222 is used to be obtained for described in characterizing according to the main characteristic vector of pixel
The accumulation direction value of the main characteristic vector of each pixel of the accumulated change degree of smooth structure tensor field;Degrees of detail obtains mould
Block 223 is used to the accumulation direction value of the main characteristic vector of all pixels point being normalized and reverse process obtains each pixel
The degrees of detail of point.
The accumulation direction value acquisition module 222 includes being used to calculate each pixel along its main characteristic vector positive direction
The positive direction accumulation direction value acquisition module 222a of accumulation direction value and for calculating each pixel along its main characteristic vector
The opposite direction accumulation direction value acquisition module 222b of the accumulation direction value of opposite direction.
Figure 12 is refer to, positive direction accumulation direction value acquisition module 222a includes:Primary vector submodule 2221a, second
Vectorial submodule 2222a, angle calcu-lation submodule 2223a, judge implementation sub-module 2224a.
Primary vector submodule 2221a is used for the pixel for setting the accumulation direction value of main characteristic vector to be calculated as first
Pixel, choose described image in be located at first pixel main characteristic vector positive direction and with the distance of first pixel
For the second pixel of unit distance, the primary vector that first pixel is constituted with second pixel is obtained.Secondary vector
Submodule 2222a be used for choose in described image positioned at second pixel main characteristic vector positive direction and with second pixel
The distance of point is the 3rd pixel of unit distance, obtains the secondary vector that second pixel is constituted with the 3rd pixel.
Angle calcu-lation submodule 2223a is used to calculate and store the angle between the primary vector and the secondary vector, and records execution
The cumulative frequency of the step.Judge that implementation sub-module 2224a is used to judge whether the cumulative frequency is less than pre-determined number, if sentenced
Disconnected result is yes, then second pixel is set as into the first pixel, the 3rd pixel is set as into the second pixel, and
Return to the secondary vector submodule 2222a, otherwise, obtain all stored angles as the pixel along its main feature to
Measure the accumulation direction value of positive direction.
Figure 13 is refer to, opposite direction accumulation direction value acquisition module 222b includes:Primary vector submodule 2221b, second
Vectorial submodule 2222b, angle calcu-lation submodule 2223b, judge implementation sub-module 2224b.
Primary vector submodule 2221b is used for the pixel for setting the accumulation direction value of main characteristic vector to be calculated as first
Pixel, choose described image in be located at first pixel main characteristic vector opposite direction and with the distance of first pixel
For the second pixel of unit distance, the primary vector that first pixel is constituted with second pixel is obtained.Secondary vector
Submodule 2222b be used for choose in described image positioned at second pixel main characteristic vector opposite direction and with second pixel
The distance of point is the 3rd pixel of unit distance, obtains the secondary vector that second pixel is constituted with the 3rd pixel.
Angle calcu-lation submodule 2223b is used to calculate and store the angle between the primary vector and the secondary vector, and records execution
The cumulative frequency of the step.Judge that implementation sub-module 2224b is used to judge whether the cumulative frequency is less than pre-determined number, if sentenced
Disconnected result is yes, then second pixel is set as into the first pixel, the 3rd pixel is set as into the second pixel, and
Return to the secondary vector submodule 2222b, otherwise, obtain all stored angles as the pixel along its main feature to
Measure the accumulation direction value of opposite direction.
In summary, relative to prior art, the present invention provide automated graphics region division realize device need not
Any user mutual can complete the region division of image, calculate relatively easy, can also be realized on the platform of poor-performing
The region division of image.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation be all between difference with other embodiment, each embodiment identical similar part mutually referring to.
For system class embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, related part is joined
See the part explanation of embodiment of the method.
It should be noted that term " comprising ", "comprising" or its any other variant are intended to the bag of nonexcludability
Contain, so that process, method, article or device including a series of key elements are not only including those key elements, but also including
Other key elements being not expressly set out, or also include for this process, method, article or the intrinsic key element of device.
In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including the key element
Process, method, article or device in also there is other identical element.
One of ordinary skill in the art will appreciate that realizing that all or part of step of above-described embodiment can be by hardware
To complete, the hardware of correlation can also be instructed to complete by program, described program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The above described is only a preferred embodiment of the present invention, any formal limitation not is made to the present invention, though
So the present invention is disclosed above with preferred embodiment, but is not limited to the present invention, any to be familiar with this professional technology people
Member, without departing from the scope of the present invention, when the technology contents using the disclosure above make a little change or modification
For the equivalent embodiment of equivalent variations, as long as being the technical spirit pair according to the present invention without departing from technical solution of the present invention content
Any simple modification, equivalent variations and modification that above example is made, in the range of still falling within technical solution of the present invention.
Claims (10)
1. a kind of implementation method of automated graphics region division, it is characterised in that methods described includes:
Obtain the smooth structure tensor field corresponding to each pixel in image;
The degrees of detail of described image is obtained according to the accumulated change degree of the smooth structure tensor field;It is described to utilize described smooth
The step of accumulated change degree of structure tensor obtains the degrees of detail of described image, including:According to the smooth of each pixel
The main characteristic vector of each pixel of structure tensor field computation;Obtain described flat for characterizing according to the main characteristic vector of pixel
The accumulation direction value of the main characteristic vector of each pixel of the accumulated change degree of Slipped Clove Hitch structure tensor field;By all pixels point
The accumulation direction value of main characteristic vector is normalized and reverse process obtains the degrees of detail of each pixel;
Region division is carried out to described image according to the degrees of detail of described image.
2. the implementation method of automated graphics region division as claimed in claim 1, it is characterised in that every in the acquisition image
The step of smooth structure tensor field corresponding to individual pixel, including:
Processing is filtered to described image, the gradient of each pixel is obtained;
According to its corresponding tensor field of the gradient calculation of each pixel;
Tensor field to each pixel is smoothed the acquisition smooth structure tensor field.
3. the implementation method of automated graphics region division as claimed in claim 1, it is characterised in that described according to pixel
Main characteristic vector obtain the main feature of each pixel of the accumulated change degree for characterizing the smooth structure tensor field to
The step of accumulation direction value of amount, including:Accumulation direction value of each pixel along its main characteristic vector positive direction is calculated respectively
And each accumulation direction value of the pixel along its main characteristic vector opposite direction.
4. the implementation method of automated graphics region division as claimed in claim 3, it is characterised in that each pixel of calculating
The step of putting along the accumulation direction value of its main characteristic vector positive direction, including:
Step a1:The pixel of accumulation direction value of main characteristic vector to be calculated is set as the first pixel, described image is chosen
In be located at first pixel main characteristic vector positive direction and with the second picture that the distance of first pixel is unit distance
Vegetarian refreshments, obtains the primary vector that first pixel is constituted with second pixel;
Step b1:Choose described image in be located at second pixel main characteristic vector positive direction and with second pixel
Distance is the 3rd pixel of unit distance, obtains the secondary vector that second pixel is constituted with the 3rd pixel;
Step c1:Calculate and store the angle between the primary vector and the secondary vector, and record the accumulation for performing the step
Number of times;
Step d1:Judge whether the cumulative frequency is less than pre-determined number, if it is judged that being yes, then set second pixel
It is set to the first pixel, the 3rd pixel is set as to the second pixel, perform step b1, otherwise, obtains all stored
Accumulation direction value of the angle as the pixel along its main characteristic vector positive direction.
5. the implementation method of automated graphics region division as claimed in claim 3, it is characterised in that each pixel of calculating
The step of putting along the accumulation direction value of its main characteristic vector opposite direction, including:
Step a2:The pixel of accumulation direction value of main characteristic vector to be calculated is set as the first pixel, described image is chosen
In be located at first pixel main characteristic vector opposite direction and with the second picture that the distance of first pixel is unit distance
Vegetarian refreshments, obtains the primary vector that first pixel is constituted with second pixel;
Step b2:Choose described image in be located at second pixel main characteristic vector opposite direction and with second pixel
Distance is the 3rd pixel of unit distance, obtains the secondary vector that second pixel is constituted with the 3rd pixel;
Step c2:Calculate and store the angle between the primary vector and the secondary vector, and record the accumulation for performing the step
Number of times;
Step d2:Judge whether the cumulative frequency is less than pre-determined number, if it is judged that being yes, then set second pixel
It is set to the first pixel, the 3rd pixel is set as to the second pixel, perform step b2, otherwise, obtains all stored
Accumulation direction value of the angle as the pixel along its main characteristic vector opposite direction.
6. a kind of automated graphics region division realizes device, it is characterised in that described device includes:
Smooth structure tensor field acquisition module, for obtaining the smooth structure tensor field in image corresponding to each pixel;
Image detail degree acquisition module, the details of described image is obtained according to the accumulated change degree of the smooth structure tensor field
Degree;
Described image degrees of detail acquisition module, including:
Main characteristic vector computing module, the main spy for each pixel of smooth structure tensor field computation according to each pixel
Levy vector;
Accumulation direction value acquisition module, for being obtained according to the main characteristic vector of pixel for characterizing the smooth structure tensor
The accumulation direction value of the main characteristic vector of each pixel of the accumulated change degree of field;
Degrees of detail acquisition module, for being normalized and reversely locating the accumulation direction value of the main characteristic vector of all pixels point
Reason obtains the degrees of detail of each pixel;
Image-region division module, region division is carried out according to the degrees of detail of described image to described image.
7. automated graphics region division as claimed in claim 6 realizes device, it is characterised in that the smooth structure tensor
Field acquisition module, including:
Gradient extraction module, for being filtered processing to described image, obtains the gradient of each pixel;
Tensor field acquisition module, for its corresponding tensor field of the gradient calculation according to each pixel;
Smoothing module, the acquisition smooth structure tensor field is smoothed for the tensor field to each pixel.
8. automated graphics region division as claimed in claim 7 realizes device, it is characterised in that the accumulation direction value obtains
Modulus block includes the positive direction accumulation direction for being used to calculate accumulation direction value of each pixel along its main characteristic vector positive direction
Value acquisition module and the opposite direction accumulation for calculating accumulation direction value of each pixel along its main characteristic vector opposite direction
Direction value acquisition module.
9. automated graphics region division as claimed in claim 8 realizes device, it is characterised in that the positive direction accumulation side
Include to value acquisition module:
Primary vector submodule:The pixel of accumulation direction value of main characteristic vector to be calculated is set as the first pixel, is chosen
In described image positioned at first pixel main characteristic vector positive direction and be unit distance with the distance of first pixel
The second pixel, obtain the primary vector that first pixel and second pixel are constituted;
Secondary vector submodule:Choose described image in be located at second pixel main characteristic vector positive direction and with this second
The distance of pixel is the 3rd pixel of unit distance, obtain that second pixel and the 3rd pixel constitute second to
Amount;
Angle calcu-lation submodule:Calculate and store the angle between the primary vector and the secondary vector, and record execution the step
Rapid cumulative frequency;
Judge implementation sub-module:Judge the cumulative frequency whether be less than pre-determined number, if it is judged that be it is yes, then by this second
Pixel is set as the first pixel, the 3rd pixel is set as to the second pixel, and returns to the secondary vector submodule
Block, otherwise, obtains accumulation direction value of all stored angles as the pixel along its main characteristic vector positive direction.
10. automated graphics region division as claimed in claim 8 realizes device, it is characterised in that the opposite direction accumulation
Direction value acquisition module includes:
Primary vector submodule:The pixel of accumulation direction value of main characteristic vector to be calculated is set as the first pixel, is chosen
In described image positioned at first pixel main characteristic vector opposite direction and be unit distance with the distance of first pixel
The second pixel, obtain the primary vector that first pixel and second pixel are constituted;
Secondary vector submodule:Choose described image in be located at second pixel main characteristic vector opposite direction and with this second
The distance of pixel is the 3rd pixel of unit distance, obtain that second pixel and the 3rd pixel constitute second to
Amount;
Angle calcu-lation submodule:Calculate and store the angle between the primary vector and the secondary vector, and record execution the step
Rapid cumulative frequency;
Judge implementation sub-module:Judge the cumulative frequency whether be less than pre-determined number, if it is judged that be it is yes, then by this second
Pixel is set as the first pixel, the 3rd pixel is set as to the second pixel, and returns to the secondary vector submodule
Block, otherwise, obtains accumulation direction value of all stored angles as the pixel along its main characteristic vector opposite direction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210560218.3A CN103886622B (en) | 2012-12-21 | 2012-12-21 | The implementation method of automated graphics region division and realize device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210560218.3A CN103886622B (en) | 2012-12-21 | 2012-12-21 | The implementation method of automated graphics region division and realize device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103886622A CN103886622A (en) | 2014-06-25 |
CN103886622B true CN103886622B (en) | 2017-10-31 |
Family
ID=50955495
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210560218.3A Active CN103886622B (en) | 2012-12-21 | 2012-12-21 | The implementation method of automated graphics region division and realize device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103886622B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104699250B (en) * | 2015-03-31 | 2018-10-19 | 小米科技有限责任公司 | Display control method and device, electronic equipment |
CN107318023B (en) * | 2017-06-21 | 2020-12-22 | 西安万像电子科技有限公司 | Image frame compression method and device |
CN108536297A (en) * | 2018-03-29 | 2018-09-14 | 北京微播视界科技有限公司 | The implementation method and device of human-computer interaction application program for more people |
CN110443907A (en) * | 2019-06-28 | 2019-11-12 | 北京市政建设集团有限责任公司 | Patrol task processing method and patrol task processing server |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102289812A (en) * | 2011-08-26 | 2011-12-21 | 上海交通大学 | Object segmentation method based on priori shape and CV (Computer Vision) model |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5505164B2 (en) * | 2010-07-23 | 2014-05-28 | ソニー株式会社 | Image processing apparatus and method, and program |
-
2012
- 2012-12-21 CN CN201210560218.3A patent/CN103886622B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102289812A (en) * | 2011-08-26 | 2011-12-21 | 上海交通大学 | Object segmentation method based on priori shape and CV (Computer Vision) model |
Non-Patent Citations (3)
Title |
---|
AN IMPROVED REGION-BASED MODEL WITH LOCAL STATISTICAL FEATURE;Qi Ge等;《2011 18th IEEE International Conference on Image Processing》;20111231;第3341-3344页 * |
基于结构张量与随机游走的图像分割算法;片兆宇 等;《东北大学学报(自然科学版)》;20090831;第30卷(第8期);第1095-1098页 * |
结合半局部信息与结构张量的无监督纹理图像分割;赵在新 等;《中国图象图形学报》;20110430;第16卷(第4期);第559-565页 * |
Also Published As
Publication number | Publication date |
---|---|
CN103886622A (en) | 2014-06-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8571271B2 (en) | Dual-phase red eye correction | |
CN103914834B (en) | A kind of significance object detecting method based on prospect priori and background priori | |
CN103745468B (en) | Significant object detecting method based on graph structure and boundary apriority | |
CN104835175B (en) | Object detection method in a kind of nuclear environment of view-based access control model attention mechanism | |
CN103886622B (en) | The implementation method of automated graphics region division and realize device | |
CN107506754A (en) | Iris identification method, device and terminal device | |
US20070036438A1 (en) | Methods and systems for identifying red eye pairs | |
US20120155756A1 (en) | Method of separating front view and background and apparatus | |
US20120070074A1 (en) | Method and System for Training a Landmark Detector using Multiple Instance Learning | |
CN101625760A (en) | Method for correcting certificate image inclination | |
CN109348731A (en) | A kind of method and device of images match | |
US9600888B2 (en) | Image processing device, image processing method, and program | |
US8027978B2 (en) | Image search method, apparatus, and program | |
CN105426828A (en) | Face detection method, face detection device and face detection system | |
Galsgaard et al. | Circular hough transform and local circularity measure for weight estimation of a graph-cut based wood stack measurement | |
US20150262362A1 (en) | Image Processor Comprising Gesture Recognition System with Hand Pose Matching Based on Contour Features | |
CN111553914A (en) | Vision-based goods detection method and device, terminal and readable storage medium | |
Malik et al. | Comparative analysis of edge detection between gray scale and color image | |
US20160253581A1 (en) | Processing system, processing method, and recording medium | |
CN101430789A (en) | Image edge detection method based on Fast Slant Stack transformation | |
CN107818552A (en) | A kind of binocular image goes reflective method | |
Chu et al. | Subimage cosegmentation in a single white blood cell image | |
US20210216829A1 (en) | Object likelihood estimation device, method, and program | |
Camaro et al. | Appearance shock grammar for fast medial axis extraction from real images | |
KR20150017762A (en) | Discrimination container generation device and pattern detection device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |